Llama CPP as encoder #426
fabiomatricardi
started this conversation in
Ideas
Llama CPP as encoder
#426
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I really like the repo!
My idea/suggestion is to use GGUF versions also of the encoders/embeddings. In fact llama-cpp-python has full classes for embeddings model too like all-MiniLM-L6-v2...
Is it already a feature available?
It would be great to no need all the time 2Gb of pytorch dependencies in a venv.
thanks
Beta Was this translation helpful? Give feedback.
All reactions